Tribotronic Transistors And Their Applications In Logic,Memory,and Neuromorphic Computing

The rapid development of the Internet of Things（Io Ts）and artificial intelligence（AI）technologies will profoundly change our way of life and work.Distributed electronic devices,serving as perception nodes for Io Ts and AI,act as bridges connecting the physical and digital worlds.To ensure these devices can operate sustainably and interact efficiently,while also guaranteeing seamless communication and intelligent interaction between users and devices,achieving low cost,multifunctionality,high energy efficiency,and wireless communication capability is paramount.Therefore,the development of integrated intelligent multifunctional electronic devices with innovative human-machine interaction technology is particularly necessary.Tribotronics,by cleverly combining the triboelectric effect with semiconductor technology,has opened up a new way of interaction,enabling the conversion of external mechanical signals into electrical signals to drive semiconductor devices.At present,various functional devices have been developed in the field of tribotronics,such as smart skins,logic circuits,electromechanical memory,tactile imaging,distance sensing,image recognition,and gas sensing,demonstrating the broad application potential of this technology.However,the performance and multifunctionality of these devices still require optimization,necessitating further development and exploration in application domains.Organic semiconductor materials have garnered significant attention due to their high performance and suitability for large-area fabrication,while two-dimensional materials with excellent electrical properties hold great potential in electronics,optoelectronics,and sensing applications.Therefore,this paper is primarily dedicated to the design and development of a series of post-Moore era electronic devices based on triboelectric effects and organic/two-dimensional semiconductor materials,featuring self-driving and active mechanical sensing capabilities.The main research content of this thesis includes the following:Firstly,a high-performance large-area tribotronic C8-BTBT-PS transistor array was developed,exhibiting an ultrahigh current on/off ratio（>10⁸）,steep subthreshold swing(29.89μm·dec^-1),high stability,and excellent reproducibility under triboelectric potential drive.Additionally,the constructed mechanically modulated tribotronic logic device demonstrated good stability and high gain(1260 V·mm^-1).The showcased large-area tribotronic C8-BTBT-PS transistor array exhibits exceptional performance,providing an effective development platform for mechanically driven electronic terminals,interactive intelligent systems,artificial machine skins,and other applications.Secondly,a triboelectric neuromorphic charge-trapping transistor based on a high-k dielectric stack gate was constructed.Low-frequency noise tests confirm that mechanical displacement modulation introduces no additional current perturbations,ensuring the reliability of the mechanical gate control technology.Excellent mechano-driven non-volatile multistate storage performance is demonstrated by modulating the amount of captured charge in the stack gate through triboelectric potential.Meanwhile,a mechanically driven programmable inverter was realized by connecting a load resistor in series.Based on the synaptic characteristics of the device,a mechano-driven artificial neural network is constructed for handwritten digit recognition,achieving an accuracy of approximately 88.59%.These results not only highlight the potential applications of triboelectric charge-trapping transistors in mechanical drive interactions,low-power data storage,and neuromorphic computing but also lay the foundation for the development of future intelligent electronic systems.Thirdly,we have successfully fabricated a self-powered reconfigurable multifunctional device by combining triboelectric nanogenerators with asymmetric van der Waals heterojunctions.On this two-dimensional heterojunction platform,polarity-switchable transistors,bipolar memory,reconfigurable diodes,logic-in-memory,and biomimetic synaptic functions have been achieved through electrostatic doping.This work demonstrates a universal device structure and operation mode applicable to all bipolar two-dimensional materials controllable via electrostatic manipulation,providing new possibilities for achieving reconfigurable multifunctional operations at the hardware level.In summary,through the coupling of high-stability triboelectric nanogenerators and functional field-effect transistors,we have successfully developed active,self-powered,high-performance,and multifunctional tribotronic devices.This article offers a new perspective on electronic devices in the post-Moore’s Law era,providing innovative ideas and technological foundations for the development of smart electronic technologies.